Systems and methods for increasing resolution of images captured by a camera sensor

Abstract

Systems, methods, and non-transitory computer readable media can obtain information relating to a bias associated with a camera sensor. A plurality of images of a scene captured by the camera sensor can be obtained, where the plurality of images are captured at a resolution supported by the camera sensor. A plurality of weights for each image of the plurality of images can be determined based at least in part on the bias. A combined image of the scene can be generated based on the plurality of images and the determined weights, the combined image having a resolution higher than the resolution supported by the camera sensor.

What is claimed is: 
     
       1. A computer-implemented method comprising:
 generating, by a computing system, a bias associated with a camera sensor using a vibrating source; 
 obtaining, by the computing system, information relating to the bias from an accelerometer associated with the camera sensor, wherein the bias includes a random bias; 
 obtaining, by the computing system, a plurality of images of a scene captured by the camera sensor, the plurality of images captured at a resolution supported by the camera sensor and aligned with respect to the camera sensor; 
 determining, by the computing system, a plurality of weights for each image of the plurality of images based at least in part on the bias; and 
 generating, by the computing system, a combined image of the scene based on the plurality of images and the determined weights, the combined image having a resolution higher than the resolution supported by the camera sensor. 
 
     
     
       2. The computer-implemented method of  claim 1 , wherein the bias provides subpixel image data for a portion of an image captured by the camera sensor. 
     
     
       3. The computer-implemented method of  claim 1 , wherein the bias further includes one or more of: a directional bias or an average bias. 
     
     
       4. The computer-implemented method of  claim 1 , wherein each image of the plurality of images includes a random set of pixels captured by the camera sensor. 
     
     
       5. The computer-implemented method of  claim 1 , wherein the plurality of weights for each image includes a weight for each pixel of the image. 
     
     
       6. The computer-implemented method of  claim 1 , wherein the plurality of weights for each image is determined based on a machine learning model. 
     
     
       7. The computer-implemented method of  claim 1 , wherein, for a particular section of the combined image, a weight assigned to a portion of an image of the plurality of images that aligns with the particular section is higher than a weight assigned to a portion of an image of the plurality of images that does not align with the particular section. 
     
     
       8. The computer-implemented method of  claim 1 , wherein a weight assigned to a portion of an image of the plurality of images that includes the bias is higher than a weight assigned to a portion of the image that does not include the bias. 
     
     
       9. The computer-implemented method of  claim 1 , wherein the generating the combined image of the scene includes performing a join on the plurality of images and the corresponding plurality of weights. 
     
     
       10. A system comprising:
 at least one hardware processor; and 
 a memory storing instructions that, when executed by the at least one processor, cause the system to perform: 
 generating a bias associated with a camera sensor using a vibrating source; 
 obtaining information relating to the bias from an accelerometer associated with the camera sensor, wherein the bias includes a random bias; 
 obtaining a plurality of images of a scene captured by the camera sensor, the plurality of images captured at a resolution supported by the camera sensor and aligned with respect to the camera sensor; 
 determining a plurality of weights for each image of the plurality of images based at least in part on the bias; and 
 generating a combined image of the scene based on the plurality of images and the determined weights, the combined image having a resolution higher than the resolution supported by the camera sensor. 
 
     
     
       11. The system of  claim 10 , wherein the bias provides subpixel image data for a portion of an image captured by the camera sensor. 
     
     
       12. The system of  claim 10 , wherein the bias further includes one or more of: a directional bias or an average bias. 
     
     
       13. The system of  claim 10 , wherein, for a particular section of the combined image, a weight assigned to a portion of an image of the plurality of images that aligns with the particular section is higher than a weight assigned to a portion of an image of the plurality of images that does not align with the particular section. 
     
     
       14. The system of  claim 10 , wherein a weight assigned to a portion of an image of the plurality of images that includes the bias is higher than a weight assigned to a portion of the image that does not include the bias. 
     
     
       15. A non-transitory computer readable medium including instructions that, when executed by at least one hardware processor of a computing system, cause the computing system to perform a method comprising:
 generating a bias associated with a camera sensor using a vibrating source; 
 obtaining information relating to the bias from an accelerometer associated with the camera sensor, wherein the bias includes a random bias; 
 obtaining a plurality of images of a scene captured by the camera sensor, the plurality of images captured at a resolution supported by the camera sensor and aligned with respect to the camera sensor; 
 determining a plurality of weights for each image of the plurality of images based at least in part on the bias; and 
 generating a combined image of the scene based on the plurality of images and the determined weights, the combined image having a resolution higher than the resolution supported by the camera sensor. 
 
     
     
       16. The non-transitory computer readable medium of  claim 15 , wherein the bias provides subpixel image data for a portion of an image captured by the camera sensor. 
     
     
       17. The non-transitory computer readable medium of  claim 15 , wherein the bias further includes one or more of: a directional bias or an average bias. 
     
     
       18. The non-transitory computer readable medium of  claim 15 , wherein, for a particular section of the combined image, a weight assigned to a portion of an image of the plurality of images that aligns with the particular section is higher than a weight assigned to a portion of an image of the plurality of images that does not align with the particular section. 
     
     
       19. The non-transitory computer readable medium of  claim 15 , wherein a weight assigned to a portion of an image of the plurality of images that includes the bias is higher than a weight assigned to a portion of the image that does not include the bias.